A wave of A.I. luminaries are sending out grim warnings about the new tech but not everyone thinks the situation is so dire

Matt O'Brien

The Associated Press

Computer scientists who helped build the foundations of today’s artificial intelligence technology are warning of its dangers, but that doesn’t mean they agree on what those dangers are or how to prevent them.<div class="paywall-selector"></div><div class="paywall">



Humanity&#8217;s survival is threatened when &#8220;smart things can outsmart us,” so-called Godfather of A.I. Geoffrey Hinton said at a conference Wednesday at the Massachusetts Institute of Technology.



After retiring from <a href="https://fortune.com/company/alphabet/" target="_blank">Google</a> so he could&nbsp;<a href="https://apnews.com/article/ai-godfather-google-geoffery-hinton-fa98c6a6fddab1d7c27560f6fcbad0ad">speak more freely</a>, the 75-year-old Hinton said he&#8217;s recently changed his views about the reasoning capabilities of the computer systems he&#8217;s spent a lifetime researching.



“These things will have learned from us, by reading all the novels that ever were and everything Machiavelli ever wrote, how to manipulate people,” Hinton said, addressing the crowd attending MIT Technology Review&#8217;s EmTech Digital conference from his home via video. “Even if they can’t directly pull levers, they can certainly get us to pull levers.”



“I wish I had a nice simple solution for this, but I don’t,” he added. “I’m not sure there is a solution.”



Fellow A.I. pioneer Yoshua Bengio, co-winner with Hinton of the <a href="https://apnews.com/article/north-america-science-technology-artificial-intelligence-nobel-prizes-078b120bec4e4a5f825821e89a749adc">top computer science prize</a>, told The Associated Press on Wednesday that he&#8217;s “pretty much aligned” with Hinton&#8217;s concerns brought on by chatbots such as ChatGPT and related technology, but worries that to simply say “We&#8217;re doomed” is not going to help.



“The main difference, I would say, is he’s kind of a pessimistic person, and I’m more on the optimistic side,” said Bengio, a professor at the University of Montreal. “I do think that the dangers—the short-term ones, the long-term ones—are very serious and need to be taken seriously by not just a few researchers but governments and the population.”



There are plenty of signs that governments are listening. The White House has called in the CEOs of Google, <a href="https://fortune.com/company/microsoft/" target="_blank">Microsoft</a> and ChatGPT-maker OpenAI to meet Thursday with Vice President Kamala Harris in what&#8217;s being described by officials as a frank discussion on how to mitigate both the near-term and long-term risks of their technology. European lawmakers are also accelerating negotiations <a href="https://apnews.com/article/technology-business-government-and-politics-artificial-intelligence-data-privacy-5226382bb316f8aad4cbc5637f03a44c">to pass sweeping new A.I. rules.</a>



But all the talk of the <a href="https://apnews.com/article/artificial-intelligence-chatgpt-risks-petition-elon-musk-steve-wozniak-534f0298d6304687ed080a5119a69962">most dire future dangers</a> has some worried that hype around superhuman machines—which don&#8217;t yet exist—is distracting from attempts to set practical safeguards on current A.I. products that are largely unregulated.



Margaret Mitchell, a former leader on Google’s A.I. ethics team, said she’s upset that Hinton didn’t speak out during his decade in a position of power at Google, especially after the 2020 ouster of prominent Black scientist Timnit Gebru, who had studied the harms of large language models before they were widely commercialized into products such as ChatGPT and Google’s Bard.



“It’s a privilege that he gets to jump from the realities of the propagation of discrimination now, the propagation of hate language, the toxicity and&nbsp;<a href="https://apnews.com/article/deepfake-porn-celebrities-dalle-stable-diffusion-midjourney-ai-e7935e9922cda82fbcfb1e1a88d9443a">nonconsensual pornography of women</a>, all of these issues that are actively harming people who are marginalized in tech,” said Mitchell, who was also forced out of Google in the aftermath of Gebru&#8217;s departure. “He’s skipping over all of those things to worry about something farther off.”



Bengio, Hinton and a third researcher, Yann LeCun, who works at Facebook parent Meta, were all awarded the Turing Prize in 2019 for their breakthroughs in the field of artificial neural networks, instrumental to the development of today&#8217;s A.I. applications such as ChatGPT.



Bengio, the only one of the three who didn&#8217;t take a job with a tech giant, has voiced concerns for years about near-term A.I. risks, including job market destabilization, automated weaponry and the dangers of biased data sets.



But those concerns have grown recently, leading Bengio to join other computer scientists and tech business leaders like Elon Musk and <a href="https://fortune.com/company/apple/" target="_blank">Apple</a> co-founder Steve Wozniak in calling for a six-month pause on developing A.I. systems more powerful than OpenAI&#8217;s latest model, GPT-4.



Bengio said Wednesday he believes the latest A.I. language models already pass the “Turing test” named after <a href="https://apnews.com/article/alan-turing-50-pound-bank-note-uk-codebreaker-d965076529b70fc8dbb020d364d88176">British codebreaker and AI pioneer Alan Turing&#8217;s method</a> introduced in 1950 to measure when A.I. becomes indistinguishable from a human — at least on the surface.



“That’s a milestone that can have drastic consequences if we’re not careful,” Bengio said. “My main concern is how they can be exploited for nefarious purposes to destabilize democracies, for cyber attacks, disinformation. You can have a conversation with these systems and think that you’re interacting with a human. They’re difficult to spot.”



Where researchers are less likely to agree is on how current A.I. language systems—which have many limitations, including a tendency to fabricate information—will actually get smarter than humans.



Aidan Gomez was one of the co-authors of the pioneering 2017 paper that introduced a so-called transformer technique—the “T” at the end of ChatGPT—for improving the performance of machine-learning systems, especially in how they learn from passages of text. Then just a 20-year-old intern at Google, Gomez remembers laying on a couch at the company&#8217;s California headquarters when his team sent out the paper around 3 a.m. when it was due.



“Aidan, this is going to be so huge,” he remembers a colleague telling him, of the work that&#8217;s since helped lead to new systems that can generate humanlike prose and imagery.



Six years later and now CEO of his own A.I. company called Cohere, which Hinton has invested in, Gomez is enthused about the potential applications of these systems but bothered by fearmongering he says is “detached from the reality” of their true capabilities and “relies on extraordinary leaps of imagination and reasoning.&#8221;



“The notion that these models are somehow gonna get access to our nuclear weapons and launch some sort of extinction-level event is not a productive discourse to have,” Gomez said. &#8220;It’s harmful to those real pragmatic policy efforts that are trying to do something good.”</div>

Humanity's survival is threatened when "smart things can outsmart us,” so-called Godfather of AI Geoffrey Hinton said at a conference Wednesday.

Tech luminaries are mixed on dire warnings

Did your workplace make our list of the 100 Best Companies to Work For?
<a href="https://fortune.com/ranking/best-companies/?&itm_source=fortune&itm_medium=article_tout&itm_campaign=AprMayMag">Explore this year's list.</a>

F-Wire

Tech

A wave of AI luminaries are sending out grim warnings about the new tech but not everyone thinks the situation is so dire

Tech luminaries are mixed on dire warnings | Fortune

Jason Del Rey

Exclusive: Chinese suppliers are offering U.S. Amazon sellers a tariff solution—but it’s not legal

/2025/04/12/china-suppliers-amazon-sellers-tariff-workaround-proposal-lowers-declared-value-goods/

James Morris

China’s EV giants shift gears toward the U.K. as tariff troubles mount in the U.S. and Europe

/europe/2025/04/12/chinas-ev-giants-shift-gears-toward-the-u-k-as-tariff-troubles-mount-in-the-u-s-and-europe/

Amanda Gerut

US President Donald Trump, Senator Dave McCormick, his wife Dina Powell McCormick and Elon Musk watch the men's NCAA wrestling competition at the Wells Fargo Center in Philadelphia, Pennsylvania, on March 22, 2025.

Mark Zuckerberg’s Meta Platforms adds former Trump advisor to the board days before an antitrust showdown with the FTC

/2025/04/12/mark-zuckerberg-meta-platforms-board-trump-dana-white-dina-powell-mccormick-antitrust-ftc/

Stuart Dyos

Intel’s new CEO invested at least $200 million in Chinese businesses, some with links to the country’s military, report says

/2025/04/12/intel-ceo-chief-executive-lip-bu-tan-china-walden-international-sakarya-seine-department-defense/

Elon Musk’s X is under investigation by Ireland’s privacy watchdog over its use of personal data to train Grok AI

/2025/04/11/elon-musk-gronk-ai-x-investigation-ireland-privacy-watchdog-personal-data/

Sharon Goldman

12 former OpenAI employees asked to be heard in Elon Musk’s lawsuit against the company; one calls Sam Altman a ‘person of low integrity’

/2025/04/11/12-ex-openai-employees-elon-musk-lawsuit-amicus-filing-lessig/

Marco Quiroz-Gutierrez

Apple airlifted 600 tons of iPhones to the U.S. to try to beat Trump’s tariffs as customers scramble to secure devices before possible price increases

https://fortune.com/2025/04/10/apple-airlifts-600-tons-iphones-india-china-tariffs/?itm_source=parsely-api

A 92-year-old refuses to sell her home to the Masters golf course that’s spent $200 million on expansion: ‘Money ain’t everything’

https://fortune.com/2025/04/12/homeowner-wont-sell-augusta-national-masters-tournament-georgia/?itm_source=parsely-api

Chris Morris

A tech CEO has been charged with fraud for saying his e-commerce startup was powered by AI, when it was actually just using manual human labor

https://fortune.com/2025/04/11/albert-saniger-nate-shopping-app-fraud-ai-justice-department/?itm_source=parsely-api

Alena Botros

Ray Dalio says a recession is likely—but not a normal one: 'We are changing the monetary order'

https://fortune.com/2025/04/11/ray-dalio-recession-changing-monetary-order/?itm_source=parsely-api

Beatrice Nolan

Elon Musk drastically drops DOGE’s savings goal from $2 trillion to $150 billion for the year

https://fortune.com/2025/04/11/elon-musk-drastically-drops-doge-savings-goal-2-trillion-150-billion-year/?itm_source=parsely-api

Exclusive: Chinese suppliers are offering U.S. Amazon sellers a tariff solution—but it's not legal

https://fortune.com/2025/04/12/china-suppliers-amazon-sellers-tariff-workaround-proposal-lowers-declared-value-goods/?itm_source=parsely-api